Integration of Experiments
Rebecca Regina Snabel
Setting Parameters:
Preprocessing performed with Seurat and Scater. Output to that Rmd file, will visualize multiple settings for PCs to use.
## Parameters for processing dataset and metadata ##
## Dimensionality reduction ##
# For PCA to run on
pcs_max = 70
# PCs used for different UMAP representations
pcs_for_overview = c(5, 9, 10, 12, 16, 18, 20, 22, 25, 30, 35, 40, 43, 45, 50, 55)
lab_col = "Library"
# Chronological order of levels (for instance timepoints)
reorder_label <- "Timepoint"
order_levels <- c("d4" ,"d5","d7", "d8","d9", "d10","d11", "d12", "d13", "d14", "d18", "d21", "m5", "m11", "m12")
pc_set = 43
res_set = 2.3
umap_col = "Experiment"
umap_col2 = "Timepoint"
label.vector = c("Experiment", "Timepoint", "Lineage")
## Storing results ##
dataset_location <- "../scrna/CMdiff_mon-EB-EHT_combined/rerun_v4/preprocess_R/20211020_results_complete_regressed/seusetv3_scnormHVG_velocity.rds"
workdir <- "../scrna/CMdiff_mon-EB-EHT_combined/rerun_v4/preprocess_R/"
# if regression is performed: this will already be included in the folder name
result_descript = "_ExpIntegration"system(paste("mkdir -p ", workdir))
knitr::opts_knit$set(root.dir = normalizePath(workdir))Loading the dataset
Normalized, scaled dataset with information on the highest variable genes (HVGs).
seuset <- readRDS(paste0(dataset_location))Dataset specific edit:
experiment_table <- read.table("../scrna/CMdiff_mon-EB-EHT_combined/sample-experiment_table.txt", header = TRUE, row.names = 1)
seuset@meta.data$Experiment <- experiment_table[seuset@meta.data$combined_id,]Setting up results directory
dateoftoday <- gsub("-", "", as.character(Sys.Date()))
resultsdir <- paste(workdir, dateoftoday, result_descript, sep = "")
system(paste("mkdir -p ", resultsdir))knitr::opts_knit$set(root.dir = normalizePath(resultsdir))Results will be stored in: resultsdir
# split dataset into a list of seurat objects
seuset.list <- SplitObject(seuset, split.by = "Experiment")# normalize and identify variable features for each dataset independently
seuset.list <- lapply(X = seuset.list, FUN = function(x) {
x <- NormalizeData(x)
x <- FindVariableFeatures(x, selection.method = "vst", nfeatures = 2000)
})# select features that are repeatedly variable across datasets for integration
features <- SelectIntegrationFeatures(object.list = seuset.list)
seuset.anchors <- FindIntegrationAnchors(object.list = seuset.list, anchor.features = features)## Scaling features for provided objects## Finding all pairwise anchors## Running CCA## Merging objects## Finding neighborhoods## Finding anchors## Found 6622 anchors## Filtering anchors## Retained 4189 anchors## Running CCA## Merging objects## Finding neighborhoods## Finding anchors## Found 2554 anchors## Filtering anchors## Retained 1784 anchors## Running CCA## Merging objects## Finding neighborhoods## Finding anchors## Found 2729 anchors## Filtering anchors## Retained 2131 anchors## Running CCA## Merging objects## Finding neighborhoods## Finding anchors## Found 6566 anchors## Filtering anchors## Retained 2782 anchors## Running CCA## Merging objects## Finding neighborhoods## Finding anchors## Found 10052 anchors## Filtering anchors## Retained 4989 anchors## Running CCA## Merging objects## Finding neighborhoods## Finding anchors## Found 2707 anchors## Filtering anchors## Retained 679 anchorsseuset.combined <- IntegrateData(anchorset = seuset.anchors)## Merging dataset 3 into 2## Extracting anchors for merged samples## Finding integration vectors## Finding integration vector weights## Integrating data## Merging dataset 1 into 2 3## Extracting anchors for merged samples## Finding integration vectors## Finding integration vector weights## Integrating data## Merging dataset 4 into 2 3 1## Extracting anchors for merged samples## Finding integration vectors## Finding integration vector weights## Integrating data# specify that we will perform downstream analysis on the corrected data note that the original
# unmodified data still resides in the 'RNA' assay
DefaultAssay(seuset.combined) <- "integrated"
# Run the standard workflow for visualization and clustering
seuset.combined <- ScaleData(seuset.combined, verbose = FALSE)
seuset.combined <- RunPCA(seuset.combined, npcs = 70, verbose = FALSE)ElbowPlot(seuset.combined, ndims = pcs_max)
seuset.combined <- RunUMAP(seuset.combined, reduction = "pca", dims = 1:30)## 17:19:37 UMAP embedding parameters a = 0.9922 b = 1.112## 17:19:37 Read 12706 rows and found 30 numeric columns## 17:19:37 Using Annoy for neighbor search, n_neighbors = 30## 17:19:37 Building Annoy index with metric = cosine, n_trees = 50## 0% 10 20 30 40 50 60 70 80 90 100%## [----|----|----|----|----|----|----|----|----|----|## **************************************************|
## 17:19:38 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b40189c871b
## 17:19:38 Searching Annoy index using 1 thread, search_k = 3000
## 17:19:41 Annoy recall = 100%
## 17:19:41 Commencing smooth kNN distance calibration using 1 thread
## 17:19:42 Initializing from normalized Laplacian + noise
## 17:19:42 Commencing optimization for 200 epochs, with 541210 positive edges
## 17:19:48 Optimization finishedseuset.combined <- FindNeighbors(seuset.combined, reduction = "pca", dims = 1:30)## Computing nearest neighbor graph
## Computing SNNseuset.combined <- FindClusters(seuset.combined, resolution = 1.5)## Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck
##
## Number of nodes: 12706
## Number of edges: 489989
##
## Running Louvain algorithm...
## Maximum modularity in 10 random starts: 0.8571
## Number of communities: 32
## Elapsed time: 1 secondsAdjusting set-up of the object
seuset.combined@meta.data[[reorder_label]] <- factor(seuset.combined@meta.data[[reorder_label]], levels = order_levels)# Visualization
DimPlot(seuset.combined, reduction = "umap", group.by = "Timepoint")
DimPlot(seuset.combined, reduction = "umap", label = TRUE, repel = TRUE)
DimPlot(seuset.combined, reduction = "umap", group.by = "Experiment")
DimPlot(seuset.combined, reduction = "umap", group.by = "Lineage")
DimPlot(seuset.combined, reduction = "umap", group.by = "Method")
FeaturePlot(seuset.combined, c("COL3A1", "MYL7", "MYL2"))
Overview of different UMAPs with varying dimensional input
seuset <- seuset.combinedVisualizing PCA results:
#PrintPCA(object = seuset.scnorm, pcs.print = 1:5, genes.print = 5, use.full = FALSE)
VizDimLoadings(object = seuset, dims = 1:10, reduction = "pca")
VizDimLoadings(object = seuset, dims = 10:20, reduction = "pca")
pdf(paste0("VizPCAplot_PCs1-", pcs_max, ".pdf"), width = 20, height = 60)
VizDimLoadings(object = seuset, dims = 1:pcs_max, reduction = "pca")
dev.off()## png
## 2DimPlot(object = seuset, reduction = "pca", group.by = lab_col)
# Helping in choosing the PCs to include in the analysis
DimHeatmap(
object = seuset,
dims = 1:5,
cells = 500,
balanced = TRUE
)
pdf(paste0("PCheatmap_PCs1-", pcs_max, ".pdf"), width = 20, height = 60)
DimHeatmap(
object = seuset,
dims = 1:pcs_max,
cells = 500,
balanced = TRUE
)
dev.off()## png
## 2Overview of different UMAPs with varying dimensional input
The combined UMAP overview with various PC inputs (1-X, with X as each of the values in the defined pcs_for_overview variable), for a first insight into the variability of the dataset.
combine_umap_plot <- function(umap_label, pcs_for_overview){
plot.list <- list()
for (i in (1:length(pcs_for_overview))){
seuset <- RunUMAP(seuset, dims = 1:pcs_for_overview[i])
dimnr <- as.character(pcs_for_overview[i])
print(dimnr)
if (i == 1){
plot.list[[dimnr]] <-
DimPlot(seuset,
reduction = "umap",
group.by = umap_label,
combine = TRUE) + ggtitle(paste0("UMAP 1:", dimnr))
} else {
plot.list[[dimnr]] <-
DimPlot(seuset,
reduction = "umap",
group.by = umap_label,
combine = TRUE) + ggtitle(paste0("UMAP 1:", dimnr)) + theme(legend.position = "none")
}
}
#Generate combined plot for umap variabel
return(plot.list)
}
#Apply to each defined umap component in label vector
for (umap_label in label.vector) {
plot.list <- combine_umap_plot(umap_label, pcs_for_overview)
pdf(
paste0("UMAPdiffsettings_",
paste(as.character(pcs_for_overview), collapse = "-"),
"_",
umap_label,
".pdf"
),
width = 20,
height = 15
)
print(CombinePlots(plot.list, nrows = round(length(pcs_for_overview)/3)))
dev.off()
}## 17:20:09 UMAP embedding parameters a = 0.9922 b = 1.112## 17:20:09 Read 12706 rows and found 5 numeric columns## 17:20:09 Using Annoy for neighbor search, n_neighbors = 30## 17:20:09 Building Annoy index with metric = cosine, n_trees = 50## 0% 10 20 30 40 50 60 70 80 90 100%## [----|----|----|----|----|----|----|----|----|----|## **************************************************|
## 17:20:10 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b407875b5f1
## 17:20:10 Searching Annoy index using 1 thread, search_k = 3000
## 17:20:15 Annoy recall = 100%
## 17:20:15 Commencing smooth kNN distance calibration using 1 thread
## 17:20:16 Initializing from normalized Laplacian + noise
## 17:20:17 Commencing optimization for 200 epochs, with 473410 positive edges
## 17:20:21 Optimization finished## [1] "5"## 17:20:22 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:20:22 Read 12706 rows and found 9 numeric columns
## 17:20:22 Using Annoy for neighbor search, n_neighbors = 30
## 17:20:22 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:20:23 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4047ea0f84
## 17:20:23 Searching Annoy index using 1 thread, search_k = 3000
## 17:20:27 Annoy recall = 100%
## 17:20:27 Commencing smooth kNN distance calibration using 1 thread
## 17:20:28 Initializing from normalized Laplacian + noise
## 17:20:28 Commencing optimization for 200 epochs, with 508302 positive edges
## 17:20:33 Optimization finished## [1] "9"## 17:20:33 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:20:33 Read 12706 rows and found 10 numeric columns
## 17:20:33 Using Annoy for neighbor search, n_neighbors = 30
## 17:20:33 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:20:34 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b40416f4854
## 17:20:34 Searching Annoy index using 1 thread, search_k = 3000
## 17:20:38 Annoy recall = 100%
## 17:20:39 Commencing smooth kNN distance calibration using 1 thread
## 17:20:39 Initializing from normalized Laplacian + noise
## 17:20:40 Commencing optimization for 200 epochs, with 514308 positive edges
## 17:20:45 Optimization finished## [1] "10"## 17:20:45 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:20:45 Read 12706 rows and found 12 numeric columns
## 17:20:45 Using Annoy for neighbor search, n_neighbors = 30
## 17:20:45 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:20:46 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b403a3f67c1
## 17:20:46 Searching Annoy index using 1 thread, search_k = 3000
## 17:20:50 Annoy recall = 100%
## 17:20:50 Commencing smooth kNN distance calibration using 1 thread
## 17:20:51 Initializing from normalized Laplacian + noise
## 17:20:51 Commencing optimization for 200 epochs, with 518994 positive edges
## 17:20:56 Optimization finished## [1] "12"## 17:20:56 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:20:56 Read 12706 rows and found 16 numeric columns
## 17:20:56 Using Annoy for neighbor search, n_neighbors = 30
## 17:20:56 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:20:57 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4069791641
## 17:20:57 Searching Annoy index using 1 thread, search_k = 3000
## 17:21:01 Annoy recall = 100%
## 17:21:01 Commencing smooth kNN distance calibration using 1 thread
## 17:21:02 Initializing from normalized Laplacian + noise
## 17:21:02 Commencing optimization for 200 epochs, with 526352 positive edges
## 17:21:07 Optimization finished## [1] "16"## 17:21:08 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:21:08 Read 12706 rows and found 18 numeric columns
## 17:21:08 Using Annoy for neighbor search, n_neighbors = 30
## 17:21:08 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:21:09 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b403c77e514
## 17:21:09 Searching Annoy index using 1 thread, search_k = 3000
## 17:21:12 Annoy recall = 100%
## 17:21:12 Commencing smooth kNN distance calibration using 1 thread
## 17:21:13 Initializing from normalized Laplacian + noise
## 17:21:13 Commencing optimization for 200 epochs, with 528346 positive edges
## 17:21:19 Optimization finished## [1] "18"## 17:21:19 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:21:19 Read 12706 rows and found 20 numeric columns
## 17:21:19 Using Annoy for neighbor search, n_neighbors = 30
## 17:21:19 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:21:20 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b401d4351cf
## 17:21:20 Searching Annoy index using 1 thread, search_k = 3000
## 17:21:23 Annoy recall = 100%
## 17:21:23 Commencing smooth kNN distance calibration using 1 thread
## 17:21:24 Initializing from normalized Laplacian + noise
## 17:21:24 Commencing optimization for 200 epochs, with 531882 positive edges
## 17:21:30 Optimization finished## [1] "20"## 17:21:30 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:21:30 Read 12706 rows and found 22 numeric columns
## 17:21:30 Using Annoy for neighbor search, n_neighbors = 30
## 17:21:30 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:21:31 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4028c1a0d9
## 17:21:31 Searching Annoy index using 1 thread, search_k = 3000
## 17:21:34 Annoy recall = 100%
## 17:21:34 Commencing smooth kNN distance calibration using 1 thread
## 17:21:35 Initializing from normalized Laplacian + noise
## 17:21:35 Commencing optimization for 200 epochs, with 533798 positive edges
## 17:21:41 Optimization finished## [1] "22"## 17:21:41 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:21:41 Read 12706 rows and found 25 numeric columns
## 17:21:41 Using Annoy for neighbor search, n_neighbors = 30
## 17:21:41 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:21:42 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b40738897a0
## 17:21:42 Searching Annoy index using 1 thread, search_k = 3000
## 17:21:45 Annoy recall = 100%
## 17:21:45 Commencing smooth kNN distance calibration using 1 thread
## 17:21:46 Initializing from normalized Laplacian + noise
## 17:21:46 Commencing optimization for 200 epochs, with 536852 positive edges
## 17:21:52 Optimization finished## [1] "25"## 17:21:52 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:21:52 Read 12706 rows and found 30 numeric columns
## 17:21:52 Using Annoy for neighbor search, n_neighbors = 30
## 17:21:52 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:21:53 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b405130d7c6
## 17:21:53 Searching Annoy index using 1 thread, search_k = 3000
## 17:21:56 Annoy recall = 100%
## 17:21:56 Commencing smooth kNN distance calibration using 1 thread
## 17:21:57 Initializing from normalized Laplacian + noise
## 17:21:57 Commencing optimization for 200 epochs, with 541210 positive edges
## 17:22:03 Optimization finished## [1] "30"## 17:22:03 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:22:03 Read 12706 rows and found 35 numeric columns
## 17:22:03 Using Annoy for neighbor search, n_neighbors = 30
## 17:22:03 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:22:04 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b40608f8b8b
## 17:22:04 Searching Annoy index using 1 thread, search_k = 3000
## 17:22:07 Annoy recall = 100%
## 17:22:07 Commencing smooth kNN distance calibration using 1 thread
## 17:22:08 Initializing from normalized Laplacian + noise
## 17:22:08 Commencing optimization for 200 epochs, with 544892 positive edges
## 17:22:14 Optimization finished## [1] "35"## 17:22:14 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:22:14 Read 12706 rows and found 40 numeric columns
## 17:22:14 Using Annoy for neighbor search, n_neighbors = 30
## 17:22:14 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:22:15 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b407f0427b7
## 17:22:15 Searching Annoy index using 1 thread, search_k = 3000
## 17:22:18 Annoy recall = 100%
## 17:22:18 Commencing smooth kNN distance calibration using 1 thread
## 17:22:19 Initializing from normalized Laplacian + noise
## 17:22:19 Commencing optimization for 200 epochs, with 549032 positive edges
## 17:22:24 Optimization finished## [1] "40"## 17:22:24 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:22:24 Read 12706 rows and found 43 numeric columns
## 17:22:24 Using Annoy for neighbor search, n_neighbors = 30
## 17:22:24 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:22:26 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b402ee14ad0
## 17:22:26 Searching Annoy index using 1 thread, search_k = 3000
## 17:22:29 Annoy recall = 100%
## 17:22:29 Commencing smooth kNN distance calibration using 1 thread
## 17:22:30 Initializing from normalized Laplacian + noise
## 17:22:30 Commencing optimization for 200 epochs, with 550748 positive edges
## 17:22:35 Optimization finished## [1] "43"## 17:22:36 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:22:36 Read 12706 rows and found 45 numeric columns
## 17:22:36 Using Annoy for neighbor search, n_neighbors = 30
## 17:22:36 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:22:37 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b40133fe05
## 17:22:37 Searching Annoy index using 1 thread, search_k = 3000
## 17:22:40 Annoy recall = 100%
## 17:22:40 Commencing smooth kNN distance calibration using 1 thread
## 17:22:41 Initializing from normalized Laplacian + noise
## 17:22:41 Commencing optimization for 200 epochs, with 552206 positive edges
## 17:22:46 Optimization finished## [1] "45"## 17:22:46 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:22:46 Read 12706 rows and found 50 numeric columns
## 17:22:46 Using Annoy for neighbor search, n_neighbors = 30
## 17:22:46 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:22:48 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b406604d166
## 17:22:48 Searching Annoy index using 1 thread, search_k = 3000
## 17:22:51 Annoy recall = 100%
## 17:22:51 Commencing smooth kNN distance calibration using 1 thread
## 17:22:52 Initializing from normalized Laplacian + noise
## 17:22:52 Commencing optimization for 200 epochs, with 555434 positive edges
## 17:22:57 Optimization finished## [1] "50"## 17:22:58 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:22:58 Read 12706 rows and found 55 numeric columns
## 17:22:58 Using Annoy for neighbor search, n_neighbors = 30
## 17:22:58 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:22:59 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4024730856
## 17:22:59 Searching Annoy index using 1 thread, search_k = 3000
## 17:23:02 Annoy recall = 100%
## 17:23:02 Commencing smooth kNN distance calibration using 1 thread
## 17:23:03 Initializing from normalized Laplacian + noise
## 17:23:03 Commencing optimization for 200 epochs, with 558416 positive edges
## 17:23:08 Optimization finished## [1] "55"## 17:23:15 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:23:15 Read 12706 rows and found 5 numeric columns
## 17:23:15 Using Annoy for neighbor search, n_neighbors = 30
## 17:23:15 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:23:17 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4050a0b650
## 17:23:17 Searching Annoy index using 1 thread, search_k = 3000
## 17:23:22 Annoy recall = 100%
## 17:23:22 Commencing smooth kNN distance calibration using 1 thread
## 17:23:23 Initializing from normalized Laplacian + noise
## 17:23:23 Commencing optimization for 200 epochs, with 473410 positive edges
## 17:23:28 Optimization finished## [1] "5"## 17:23:28 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:23:28 Read 12706 rows and found 9 numeric columns
## 17:23:28 Using Annoy for neighbor search, n_neighbors = 30
## 17:23:28 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:23:29 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4027295723
## 17:23:29 Searching Annoy index using 1 thread, search_k = 3000
## 17:23:33 Annoy recall = 100%
## 17:23:33 Commencing smooth kNN distance calibration using 1 thread
## 17:23:34 Initializing from normalized Laplacian + noise
## 17:23:34 Commencing optimization for 200 epochs, with 508302 positive edges
## 17:23:40 Optimization finished## [1] "9"## 17:23:40 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:23:40 Read 12706 rows and found 10 numeric columns
## 17:23:40 Using Annoy for neighbor search, n_neighbors = 30
## 17:23:40 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:23:41 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b40209e9ad8
## 17:23:41 Searching Annoy index using 1 thread, search_k = 3000
## 17:23:45 Annoy recall = 100%
## 17:23:45 Commencing smooth kNN distance calibration using 1 thread
## 17:23:46 Initializing from normalized Laplacian + noise
## 17:23:46 Commencing optimization for 200 epochs, with 514308 positive edges
## 17:23:51 Optimization finished## [1] "10"## 17:23:51 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:23:51 Read 12706 rows and found 12 numeric columns
## 17:23:51 Using Annoy for neighbor search, n_neighbors = 30
## 17:23:51 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:23:52 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b404a514e0c
## 17:23:52 Searching Annoy index using 1 thread, search_k = 3000
## 17:23:56 Annoy recall = 100%
## 17:23:56 Commencing smooth kNN distance calibration using 1 thread
## 17:23:57 Initializing from normalized Laplacian + noise
## 17:23:57 Commencing optimization for 200 epochs, with 518994 positive edges
## 17:24:02 Optimization finished## [1] "12"## 17:24:02 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:24:02 Read 12706 rows and found 16 numeric columns
## 17:24:02 Using Annoy for neighbor search, n_neighbors = 30
## 17:24:02 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:24:03 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b402be2fe28
## 17:24:03 Searching Annoy index using 1 thread, search_k = 3000
## 17:24:07 Annoy recall = 100%
## 17:24:07 Commencing smooth kNN distance calibration using 1 thread
## 17:24:08 Initializing from normalized Laplacian + noise
## 17:24:08 Commencing optimization for 200 epochs, with 526352 positive edges
## 17:24:14 Optimization finished## [1] "16"## 17:24:14 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:24:14 Read 12706 rows and found 18 numeric columns
## 17:24:14 Using Annoy for neighbor search, n_neighbors = 30
## 17:24:14 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:24:15 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4010af99b0
## 17:24:15 Searching Annoy index using 1 thread, search_k = 3000
## 17:24:18 Annoy recall = 100%
## 17:24:18 Commencing smooth kNN distance calibration using 1 thread
## 17:24:19 Initializing from normalized Laplacian + noise
## 17:24:19 Commencing optimization for 200 epochs, with 528346 positive edges
## 17:24:25 Optimization finished## [1] "18"## 17:24:25 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:24:25 Read 12706 rows and found 20 numeric columns
## 17:24:25 Using Annoy for neighbor search, n_neighbors = 30
## 17:24:25 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:24:26 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4062edd527
## 17:24:26 Searching Annoy index using 1 thread, search_k = 3000
## 17:24:29 Annoy recall = 100%
## 17:24:29 Commencing smooth kNN distance calibration using 1 thread
## 17:24:30 Initializing from normalized Laplacian + noise
## 17:24:30 Commencing optimization for 200 epochs, with 531882 positive edges
## 17:24:36 Optimization finished## [1] "20"## 17:24:36 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:24:36 Read 12706 rows and found 22 numeric columns
## 17:24:36 Using Annoy for neighbor search, n_neighbors = 30
## 17:24:36 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:24:37 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4037233334
## 17:24:37 Searching Annoy index using 1 thread, search_k = 3000
## 17:24:40 Annoy recall = 100%
## 17:24:40 Commencing smooth kNN distance calibration using 1 thread
## 17:24:41 Initializing from normalized Laplacian + noise
## 17:24:41 Commencing optimization for 200 epochs, with 533798 positive edges
## 17:24:47 Optimization finished## [1] "22"## 17:24:47 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:24:47 Read 12706 rows and found 25 numeric columns
## 17:24:47 Using Annoy for neighbor search, n_neighbors = 30
## 17:24:47 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:24:48 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4032cd1cae
## 17:24:48 Searching Annoy index using 1 thread, search_k = 3000
## 17:24:51 Annoy recall = 100%
## 17:24:51 Commencing smooth kNN distance calibration using 1 thread
## 17:24:52 Initializing from normalized Laplacian + noise
## 17:24:52 Commencing optimization for 200 epochs, with 536852 positive edges
## 17:24:58 Optimization finished## [1] "25"## 17:24:58 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:24:58 Read 12706 rows and found 30 numeric columns
## 17:24:58 Using Annoy for neighbor search, n_neighbors = 30
## 17:24:58 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:24:59 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b407e3db2d6
## 17:24:59 Searching Annoy index using 1 thread, search_k = 3000
## 17:25:02 Annoy recall = 100%
## 17:25:02 Commencing smooth kNN distance calibration using 1 thread
## 17:25:03 Initializing from normalized Laplacian + noise
## 17:25:03 Commencing optimization for 200 epochs, with 541210 positive edges
## 17:25:09 Optimization finished## [1] "30"## 17:25:09 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:25:09 Read 12706 rows and found 35 numeric columns
## 17:25:09 Using Annoy for neighbor search, n_neighbors = 30
## 17:25:09 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:25:10 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b403a6c1e56
## 17:25:10 Searching Annoy index using 1 thread, search_k = 3000
## 17:25:13 Annoy recall = 100%
## 17:25:13 Commencing smooth kNN distance calibration using 1 thread
## 17:25:14 Initializing from normalized Laplacian + noise
## 17:25:14 Commencing optimization for 200 epochs, with 544892 positive edges
## 17:25:20 Optimization finished## [1] "35"## 17:25:20 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:25:20 Read 12706 rows and found 40 numeric columns
## 17:25:20 Using Annoy for neighbor search, n_neighbors = 30
## 17:25:20 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:25:21 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b401368be22
## 17:25:21 Searching Annoy index using 1 thread, search_k = 3000
## 17:25:24 Annoy recall = 100%
## 17:25:24 Commencing smooth kNN distance calibration using 1 thread
## 17:25:25 Initializing from normalized Laplacian + noise
## 17:25:25 Commencing optimization for 200 epochs, with 549032 positive edges
## 17:25:31 Optimization finished## [1] "40"## 17:25:31 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:25:31 Read 12706 rows and found 43 numeric columns
## 17:25:31 Using Annoy for neighbor search, n_neighbors = 30
## 17:25:31 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:25:32 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b402748881
## 17:25:32 Searching Annoy index using 1 thread, search_k = 3000
## 17:25:35 Annoy recall = 100%
## 17:25:35 Commencing smooth kNN distance calibration using 1 thread
## 17:25:36 Initializing from normalized Laplacian + noise
## 17:25:36 Commencing optimization for 200 epochs, with 550748 positive edges
## 17:25:42 Optimization finished## [1] "43"## 17:25:42 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:25:42 Read 12706 rows and found 45 numeric columns
## 17:25:42 Using Annoy for neighbor search, n_neighbors = 30
## 17:25:42 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:25:43 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b40554f1773
## 17:25:43 Searching Annoy index using 1 thread, search_k = 3000
## 17:25:46 Annoy recall = 100%
## 17:25:46 Commencing smooth kNN distance calibration using 1 thread
## 17:25:47 Initializing from normalized Laplacian + noise
## 17:25:47 Commencing optimization for 200 epochs, with 552206 positive edges
## 17:25:53 Optimization finished## [1] "45"## 17:25:53 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:25:53 Read 12706 rows and found 50 numeric columns
## 17:25:53 Using Annoy for neighbor search, n_neighbors = 30
## 17:25:53 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:25:54 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b40bde7413
## 17:25:54 Searching Annoy index using 1 thread, search_k = 3000
## 17:25:57 Annoy recall = 100%
## 17:25:58 Commencing smooth kNN distance calibration using 1 thread
## 17:25:58 Initializing from normalized Laplacian + noise
## 17:25:58 Commencing optimization for 200 epochs, with 555434 positive edges
## 17:26:04 Optimization finished## [1] "50"## 17:26:04 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:26:04 Read 12706 rows and found 55 numeric columns
## 17:26:04 Using Annoy for neighbor search, n_neighbors = 30
## 17:26:04 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:26:05 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b404a5e9805
## 17:26:05 Searching Annoy index using 1 thread, search_k = 3000
## 17:26:08 Annoy recall = 100%
## 17:26:09 Commencing smooth kNN distance calibration using 1 thread
## 17:26:09 Initializing from normalized Laplacian + noise
## 17:26:10 Commencing optimization for 200 epochs, with 558416 positive edges
## 17:26:15 Optimization finished## [1] "55"## 17:26:22 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:26:22 Read 12706 rows and found 5 numeric columns
## 17:26:22 Using Annoy for neighbor search, n_neighbors = 30
## 17:26:22 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:26:24 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b40461ddbd5
## 17:26:24 Searching Annoy index using 1 thread, search_k = 3000
## 17:26:28 Annoy recall = 100%
## 17:26:29 Commencing smooth kNN distance calibration using 1 thread
## 17:26:29 Initializing from normalized Laplacian + noise
## 17:26:30 Commencing optimization for 200 epochs, with 473410 positive edges
## 17:26:35 Optimization finished## [1] "5"## 17:26:35 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:26:35 Read 12706 rows and found 9 numeric columns
## 17:26:35 Using Annoy for neighbor search, n_neighbors = 30
## 17:26:35 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:26:36 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4033d7ae46
## 17:26:36 Searching Annoy index using 1 thread, search_k = 3000
## 17:26:40 Annoy recall = 100%
## 17:26:40 Commencing smooth kNN distance calibration using 1 thread
## 17:26:41 Initializing from normalized Laplacian + noise
## 17:26:41 Commencing optimization for 200 epochs, with 508302 positive edges
## 17:26:46 Optimization finished## [1] "9"## 17:26:46 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:26:46 Read 12706 rows and found 10 numeric columns
## 17:26:46 Using Annoy for neighbor search, n_neighbors = 30
## 17:26:46 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:26:47 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b40533644dc
## 17:26:47 Searching Annoy index using 1 thread, search_k = 3000
## 17:26:52 Annoy recall = 100%
## 17:26:52 Commencing smooth kNN distance calibration using 1 thread
## 17:26:52 Initializing from normalized Laplacian + noise
## 17:26:53 Commencing optimization for 200 epochs, with 514308 positive edges
## 17:26:58 Optimization finished## [1] "10"## 17:26:58 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:26:58 Read 12706 rows and found 12 numeric columns
## 17:26:58 Using Annoy for neighbor search, n_neighbors = 30
## 17:26:58 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:26:59 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4063612da4
## 17:26:59 Searching Annoy index using 1 thread, search_k = 3000
## 17:27:03 Annoy recall = 100%
## 17:27:03 Commencing smooth kNN distance calibration using 1 thread
## 17:27:04 Initializing from normalized Laplacian + noise
## 17:27:04 Commencing optimization for 200 epochs, with 518994 positive edges
## 17:27:09 Optimization finished## [1] "12"## 17:27:09 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:27:09 Read 12706 rows and found 16 numeric columns
## 17:27:09 Using Annoy for neighbor search, n_neighbors = 30
## 17:27:09 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:27:10 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b405c994f1f
## 17:27:10 Searching Annoy index using 1 thread, search_k = 3000
## 17:27:14 Annoy recall = 100%
## 17:27:14 Commencing smooth kNN distance calibration using 1 thread
## 17:27:15 Initializing from normalized Laplacian + noise
## 17:27:15 Commencing optimization for 200 epochs, with 526352 positive edges
## 17:27:20 Optimization finished## [1] "16"## 17:27:21 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:27:21 Read 12706 rows and found 18 numeric columns
## 17:27:21 Using Annoy for neighbor search, n_neighbors = 30
## 17:27:21 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:27:21 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4046bedc7c
## 17:27:22 Searching Annoy index using 1 thread, search_k = 3000
## 17:27:25 Annoy recall = 100%
## 17:27:25 Commencing smooth kNN distance calibration using 1 thread
## 17:27:26 Initializing from normalized Laplacian + noise
## 17:27:26 Commencing optimization for 200 epochs, with 528346 positive edges
## 17:27:32 Optimization finished## [1] "18"## 17:27:32 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:27:32 Read 12706 rows and found 20 numeric columns
## 17:27:32 Using Annoy for neighbor search, n_neighbors = 30
## 17:27:32 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:27:33 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b403492056a
## 17:27:33 Searching Annoy index using 1 thread, search_k = 3000
## 17:27:36 Annoy recall = 100%
## 17:27:36 Commencing smooth kNN distance calibration using 1 thread
## 17:27:37 Initializing from normalized Laplacian + noise
## 17:27:37 Commencing optimization for 200 epochs, with 531882 positive edges
## 17:27:43 Optimization finished## [1] "20"## 17:27:43 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:27:43 Read 12706 rows and found 22 numeric columns
## 17:27:43 Using Annoy for neighbor search, n_neighbors = 30
## 17:27:43 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:27:44 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b403d28daaa
## 17:27:44 Searching Annoy index using 1 thread, search_k = 3000
## 17:27:47 Annoy recall = 100%
## 17:27:47 Commencing smooth kNN distance calibration using 1 thread
## 17:27:48 Initializing from normalized Laplacian + noise
## 17:27:48 Commencing optimization for 200 epochs, with 533798 positive edges
## 17:27:54 Optimization finished## [1] "22"## 17:27:54 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:27:54 Read 12706 rows and found 25 numeric columns
## 17:27:54 Using Annoy for neighbor search, n_neighbors = 30
## 17:27:54 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:27:55 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4045c30434
## 17:27:55 Searching Annoy index using 1 thread, search_k = 3000
## 17:27:58 Annoy recall = 100%
## 17:27:58 Commencing smooth kNN distance calibration using 1 thread
## 17:27:59 Initializing from normalized Laplacian + noise
## 17:27:59 Commencing optimization for 200 epochs, with 536852 positive edges
## 17:28:05 Optimization finished## [1] "25"## 17:28:05 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:28:05 Read 12706 rows and found 30 numeric columns
## 17:28:05 Using Annoy for neighbor search, n_neighbors = 30
## 17:28:05 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:28:06 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b406373503a
## 17:28:06 Searching Annoy index using 1 thread, search_k = 3000
## 17:28:09 Annoy recall = 100%
## 17:28:09 Commencing smooth kNN distance calibration using 1 thread
## 17:28:10 Initializing from normalized Laplacian + noise
## 17:28:10 Commencing optimization for 200 epochs, with 541210 positive edges
## 17:28:16 Optimization finished## [1] "30"## 17:28:16 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:28:16 Read 12706 rows and found 35 numeric columns
## 17:28:16 Using Annoy for neighbor search, n_neighbors = 30
## 17:28:16 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:28:17 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b403e5cd8af
## 17:28:17 Searching Annoy index using 1 thread, search_k = 3000
## 17:28:20 Annoy recall = 100%
## 17:28:20 Commencing smooth kNN distance calibration using 1 thread
## 17:28:21 Initializing from normalized Laplacian + noise
## 17:28:21 Commencing optimization for 200 epochs, with 544892 positive edges
## 17:28:27 Optimization finished## [1] "35"## 17:28:27 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:28:27 Read 12706 rows and found 40 numeric columns
## 17:28:27 Using Annoy for neighbor search, n_neighbors = 30
## 17:28:27 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:28:28 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b402bc7d59a
## 17:28:28 Searching Annoy index using 1 thread, search_k = 3000
## 17:28:31 Annoy recall = 100%
## 17:28:31 Commencing smooth kNN distance calibration using 1 thread
## 17:28:32 Initializing from normalized Laplacian + noise
## 17:28:32 Commencing optimization for 200 epochs, with 549032 positive edges
## 17:28:38 Optimization finished## [1] "40"## 17:28:38 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:28:38 Read 12706 rows and found 43 numeric columns
## 17:28:38 Using Annoy for neighbor search, n_neighbors = 30
## 17:28:38 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:28:39 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b407e65890
## 17:28:39 Searching Annoy index using 1 thread, search_k = 3000
## 17:28:42 Annoy recall = 100%
## 17:28:42 Commencing smooth kNN distance calibration using 1 thread
## 17:28:43 Initializing from normalized Laplacian + noise
## 17:28:43 Commencing optimization for 200 epochs, with 550748 positive edges
## 17:28:49 Optimization finished## [1] "43"## 17:28:49 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:28:49 Read 12706 rows and found 45 numeric columns
## 17:28:49 Using Annoy for neighbor search, n_neighbors = 30
## 17:28:49 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:28:50 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b4072d8b6c5
## 17:28:50 Searching Annoy index using 1 thread, search_k = 3000
## 17:28:53 Annoy recall = 100%
## 17:28:53 Commencing smooth kNN distance calibration using 1 thread
## 17:28:54 Initializing from normalized Laplacian + noise
## 17:28:54 Commencing optimization for 200 epochs, with 552206 positive edges
## 17:29:00 Optimization finished## [1] "45"## 17:29:00 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:29:00 Read 12706 rows and found 50 numeric columns
## 17:29:00 Using Annoy for neighbor search, n_neighbors = 30
## 17:29:00 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:29:01 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b407c688bea
## 17:29:01 Searching Annoy index using 1 thread, search_k = 3000
## 17:29:04 Annoy recall = 100%
## 17:29:04 Commencing smooth kNN distance calibration using 1 thread
## 17:29:05 Initializing from normalized Laplacian + noise
## 17:29:05 Commencing optimization for 200 epochs, with 555434 positive edges
## 17:29:11 Optimization finished## [1] "50"## 17:29:11 UMAP embedding parameters a = 0.9922 b = 1.112
## 17:29:11 Read 12706 rows and found 55 numeric columns
## 17:29:11 Using Annoy for neighbor search, n_neighbors = 30
## 17:29:11 Building Annoy index with metric = cosine, n_trees = 50
## 0% 10 20 30 40 50 60 70 80 90 100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 17:29:12 Writing NN index file to temp file /scratch/snabel/tmp/BoiX/Rtmp01Usjr/file22b402f0fafb3
## 17:29:12 Searching Annoy index using 1 thread, search_k = 3000
## 17:29:15 Annoy recall = 100%
## 17:29:16 Commencing smooth kNN distance calibration using 1 thread
## 17:29:16 Initializing from normalized Laplacian + noise
## 17:29:16 Commencing optimization for 200 epochs, with 558416 positive edges
## 17:29:22 Optimization finished## [1] "55"# Saving the dataset with the normalized, scaled and identified HVGs (stored in seuset.scnorm@var.genes).
saveRDS(seuset, file="seusetv3_integrated.rds")sessionInfo()## R version 4.1.0 (2021-05-18)
## Platform: x86_64-conda-linux-gnu (64-bit)
## Running under: Ubuntu 20.04.3 LTS
##
## Matrix products: default
## BLAS/LAPACK: /vol/mbconda/snabel/anaconda3/envs/kb_scrna_R_mon3/lib/libopenblasp-r0.3.17.so
##
## locale:
## [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
## [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
## [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
## [7] LC_PAPER=en_US.UTF-8 LC_NAME=C
## [9] LC_ADDRESS=C LC_TELEPHONE=C
## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
##
## attached base packages:
## [1] stats graphics grDevices utils datasets methods base
##
## other attached packages:
## [1] patchwork_1.1.1 Matrix_1.3-4 scales_1.1.1 RColorBrewer_1.1-2
## [5] SeuratObject_4.0.2 Seurat_4.0.4 knitr_1.33 tidyr_1.1.3
## [9] dplyr_1.0.7 ggplot2_3.3.5
##
## loaded via a namespace (and not attached):
## [1] Rtsne_0.15 colorspace_2.0-2 deldir_0.2-10
## [4] ellipsis_0.3.2 ggridges_0.5.3 spatstat.data_2.1-0
## [7] farver_2.1.0 leiden_0.3.9 listenv_0.8.0
## [10] ggrepel_0.9.1 RSpectra_0.16-0 fansi_0.5.0
## [13] codetools_0.2-18 splines_4.1.0 polyclip_1.10-0
## [16] jsonlite_1.7.2 ica_1.0-2 cluster_2.1.2
## [19] png_0.1-7 uwot_0.1.10 shiny_1.6.0
## [22] sctransform_0.3.2 spatstat.sparse_2.0-0 compiler_4.1.0
## [25] httr_1.4.2 assertthat_0.2.1 fastmap_1.1.0
## [28] lazyeval_0.2.2 cli_3.0.1 limma_3.48.3
## [31] later_1.3.0 htmltools_0.5.2 tools_4.1.0
## [34] igraph_1.2.6 gtable_0.3.0 glue_1.4.2
## [37] RANN_2.6.1 reshape2_1.4.4 Rcpp_1.0.7
## [40] scattermore_0.7 jquerylib_0.1.4 vctrs_0.3.8
## [43] nlme_3.1-152 lmtest_0.9-38 xfun_0.25
## [46] stringr_1.4.0 globals_0.14.0 mime_0.11
## [49] miniUI_0.1.1.1 lifecycle_1.0.0 irlba_2.3.3
## [52] goftest_1.2-2 future_1.22.1 MASS_7.3-54
## [55] zoo_1.8-9 spatstat.core_2.3-0 promises_1.2.0.1
## [58] spatstat.utils_2.2-0 parallel_4.1.0 yaml_2.2.1
## [61] reticulate_1.20 pbapply_1.4-3 gridExtra_2.3
## [64] sass_0.4.0 rpart_4.1-15 stringi_1.7.4
## [67] highr_0.9 rlang_0.4.11 pkgconfig_2.0.3
## [70] matrixStats_0.60.1 evaluate_0.14 lattice_0.20-44
## [73] ROCR_1.0-11 purrr_0.3.4 tensor_1.5
## [76] labeling_0.4.2 htmlwidgets_1.5.3 cowplot_1.1.1
## [79] tidyselect_1.1.1 parallelly_1.27.0 RcppAnnoy_0.0.19
## [82] plyr_1.8.6 magrittr_2.0.1 R6_2.5.1
## [85] generics_0.1.0 DBI_1.1.1 pillar_1.6.2
## [88] withr_2.4.2 mgcv_1.8-36 fitdistrplus_1.1-5
## [91] survival_3.2-13 abind_1.4-5 tibble_3.1.4
## [94] future.apply_1.8.1 crayon_1.4.1 KernSmooth_2.23-20
## [97] utf8_1.2.2 spatstat.geom_2.2-2 plotly_4.9.4.1
## [100] rmarkdown_2.10 grid_4.1.0 data.table_1.14.0
## [103] digest_0.6.27 xtable_1.8-4 httpuv_1.6.2
## [106] munsell_0.5.0 viridisLite_0.4.0 bslib_0.3.0